## Tuesday, April 14, 2015

### Calculating Forcing Moves: Defensive Idea 5 - Counterattack to DRAW (stalemate)

In the last post we looked at positions where there was a clear threat on the board, and we defended by creating an equal or greater threat (EGT).  In that position we used two checks to force an equal trade of one of the pieces threatening mate, which, since we were up material, resulted in a winning position.

As noted in the last post, when we find we can create an equal or greater threat, it is often unclear where it will lead and the variations can become quite sharp and complicated.  To maintain focus through those complicated variations, I recommend looking for all three possible outcomes with your counterattack:  win, draw, or defend.

Last post we looked at positions where we were able to use an EGT to force a winning position.  In this next series of posts we’ll look at positions were we use an EGT to force positions that are draws.  So let's get started!

When it comes to drawing there are three types of positions you can try to reach:
Draw by stalemate;
Material draw (equal material, fortresses, etc); and,
Draw by repetition.

I personally start my search for draws by looking for stalemates, then for material draws.  This is because the analysis for stalemate/material draws can often be done quickly based on a material evaluation, and maybe some short calculations to see if you can forcibly trade off pieces.  Draw by repetition is often more dynamic and can involve some pretty cool positions with lots of material still left on the board.

For this post I am going to follow the same process I’ve used in previous posts for analyzing the position -- evaluate the material, identify/prioritize threats, and search for specific moves using the five defensive ideas.  But -- based on feedback from Tomasz -- I am going to try out another format for presenting the analysis.  Let me know what you think!

Black just played 1…d2.  White to move.

 8/8/8/2p5/2P5/kr6/1r1p4/K1Q5 w - - 0 1

Let's evaluate the position, and then find and prioritize all of white's threats.

Evaluation:  White is down two points (Queen vs two Rooks and a Pawn), and all of black’s pieces are very active compared to white’s.  Black is playing for a win, and white is fighting for a draw.

Threats:  Black is threatening to capture white’s queen and promote to a rook/queen giving checkmate.  If the white queen moves and unpins black’s b2 rook, black can checkmate with …Ra2#, and if the white queen leaves the defense of the first rank (b1 in particular) black will checkmate with …Rb1#.  If we skip white’s move we would quickly see that the most immediate checkmate threat is 2…dxc1=Q/R#, which is our priority for calculations.

Now that we have found black’s greatest threat, let's now look at the five defensive ideas to see what candidate moves we find, calculate each one, and then pick the best move. I am going to try this chart format to support the verbal analysis.  Hopefully this helps convey the ideas a bit better, but let me know what you think.  Take a look at this chart:

The five defensive Ideas are in the column on the left.  Then there are columns for specific moves, space to calculate those moves, enter the evaluation, and then track your King of The Hill rating.  This follows the general process of calculation – use ideas to generate specific candidate moves, calculate each one completely, then pick the best move.

We will start the process by searching for specific moves to put in the “moves” column next to each of the five ideas.

Idea 1 – do something to the attacking piece (the d2 pawn).  We can capture it (2. Qxd2), we cannot pin it to anything of greater value, and we cannot deflect it.  So Idea 1 has generated one specific candidate move.  We enter it, and move on to defensive Idea 2 to search for more candidates.

Idea 2 – block the attacking piece (the d2 pawn).  It is not possible because there are no squares between the pawn and queen. It is in fact never possible to block a pawn, knight, or king.

Idea 3 – move the piece being attacked (the white queen).  There is only square the queen can move to that keeps an eye on all three key squares (a2, b1, and d1) – that gives us 2. Qc2.  The queen has several other squares she can escape to, but they can all be refuted by black with an immediate mate in one (either Ra2#, Rb1# or d1=R/Q#).  So Idea 3 has helped us find our second specific candidate move (2. Qc2).

Idea 4 – reinforce/defend the attacked square or piece (the white queen).  There are no legal ways to defend the queen.

Idea 5 – counterattack with an equal or greater threat (of checkmate in one).  Looking at the EGT chart from earlier we can see that against a mate in one threat, the only EGT possible must begin with a check.  In this position there is only one, and that’s the capture-check on b2 (2. Qxb2+).

This process has generated three specific candidate moves that we can enter in our chart:

The next step is to calculate each of those three (and ONLY those three!!) candidate moves.  How to pick which one to calculate first?

We can either start with the most forcing, or the easiest to refute.  I think I’ll start with the most forcing this time, which is the capture-check 2. Qxb2+:

This greater threat allows us to change the position, but it is not yet clear if it will lead to a WIN, DRAW, or new possibilities to DEFEND…or if it leads to a loss and the move is just plain bad!

As mentioned above I personally like to start by looking for draws by stalemate first because it is the easiest to eliminate.  In most positions you’ll have multiple pawn moves or your pieces will enjoy freedom of movement around the board, and you can very easily see if stalemate is an option.  A quick glance at this position shows white has very limited moves, which is an indication that stalemate IS in fact a defensive option for white.  White's only pawn has no move or captures, and the king is in the corner and cannot move.  If we can force the queen off of the board then white will have no legal moves...we would just have to make sure that it is white's turn to move and not black's, which would turn a stalemate position into a checkmate position!!  Stalemate positions are often very easy to recognize.  The tricky part is to remember to consciously search for it!

Back to the check.  Black now has one capture, and two king moves in response.  Let’s see which of the three variations is best for black starting with the most forcing (the capture):

(1) 1…d2 2. Qxb2+ Rxb2  Stalemate.  Since our initial assessment was that white should be fighting for a draw, this should be good for white.  But before making this our KOTH, we need to see if there are any replies black has that are better for him than a draw?  Let’s see.

(2) 1…d2 2. Qxb2+ Kb4 3. Qxd2+ Kxc4  Endgame tablebases will say this is a win for white in like 50 moves.  The general process is to get the king active, win the black pawn, then the Queen vs Rook endgame is a technical win.  Practically speaking this endgame is very hard to actually win, but black is at a disadvantage because his king is already on the edge of the board.  Although the outcome OTB would most likely be a draw, it is only white that has any winning chances.  So black should reject this line in favor of the forced draw by stalemate.  There is only one other option for black after the capture check.

(3) 1…d2 2. Qxb2+ Ka4 3. Qxd2  This is almost identical to the variation above, except this time white has an extra pawn, so the evaluation would be slightly better for white – most likely a draw, but winning chances for white.  Black should reject this line, too, in favor of the stalemate since a draw is better than a loss!

We can now say that the mainline for this variation is the first one, leading to stalemate.  Here is an updated chart showing our variations, evaluations, and current KOTH:

White’s current KOTH is stalemate, which is pretty good considering he was fighting for a draw in our initial evaluation.  But let’s check to see if either of white’s other candidate moves offer him anything better than a draw, starting with the most forcing (the capture) 2. Qxd2:

1…d2 2. Qxd2 (unpinning the b2 rook) Rb1#  A clear refutation of white's candidate!  Here’s our updated chart showing this line:

White’s current KOTH is still stalemate, but let’s quickly check his last candidate move 2. Qc2:

This candidate is also easily refuted by black with the simple promotion -- 1…d2 2. Qc2 d1=Q+ 3. Qxd1 Ra2# (3. Qc1/Qb1 QxQ#)

Here again is an updated chart showing this final candidate move:

So this process has shown us that white has only three candidate moves to consider -- one of them leads to a stalemate by force (since all other black replies lead to winning chances for white), and the other two candidates lead directly to white being checkmated.  In conclusion, white's best move is to go for the stalemate.

This is the process you should be able to do mentally:  use ideas to search for specific candidate moves, calculate each one completely, and then pick the best variation based on the final evaluation.

In the next post we'll look at positions where we use an equal or greater threat to force a position with equal material.

1. Hello :) - Part 1

The actual format for presenting the analysis is really nice one! Well done! There is some room for improvement, but I think we will see what changes are necessary... in a practice! The first idea is to avoid presenting "empty table" due to repetition. In my opinion it would be better to leave the first table or replace the second one... in the place of this "empty" one. This way you avoid being redundant - let me know what you think about it and if you know what I mean.

There are some issues I want to address.

1. Black is playing for a win, and white is fighting for a draw. Sorry, but this sentence is NOT logical (even if repeated quite often). Much better would be to describe it in another way. For example: "Black has very active pieces and material advantage and white's king is squeezed. That's the reason why white is looking for some ways to escape from black's well placed army (pieces) that threaten to mate white's king".

2. "...if the white queen leaves the first rank black will checkmate with …Rb1#." Not precise. If White plays Qc2 (it leaves the 1st rank), black CANNOT mate with Rb1. That's the exception to your statement.

3. "Idea 2 – block the attacking piece (the d2 pawn).". I would it need some clarification. Probably you may think about this one "Idea 2 – block the attacking piece (the d2 pawn)... with any other of your pieces". Maybe it sound obvious, but I understood it like "If I move my Queen in front of the pawn - I will BLOCK it - the pawn cannot move as it is stopped/blocked".

4. "It is in fact never possible to block a pawn, knight, or king." I would expand it a little bit to this one: "It is in fact never possible to block a pawn, knight, or king... and it means you can ONLY block the linear pieces (Queen, Rook and Bishop)".

5. "This IS because the analysis for stalemate/material" ("IS" is needed after "This").

6. Find ALL "2. Qxb2" and replace to "2. Qxb2+" (plus as a check at the end is needed).

7. "A quick glance at this position shows white has very limited moves, which is an indication that stalemate IS in fact a defensive option for white.". I would DEFINITELY add some more explanation. For example: "White King and pawn has NO legal moves. If the Queen disappear and it would be Black's move - it is a great example of stalemate!". To be honest - it is the most important advice/hint (element) that I find missing in your text (article). And it is extremally practical one due to simplicity and speed of (stalemate pattern) recognition. PLEASE add it because many beginners would have some problems to "see it" clearly. Thanks in advance.

* (to be continued)...

1. Tomasz -- thanks again for your feedback. I've taken many of your suggestions to improve the post. Thank you! And for the few points I disagree with I have replied below.

I'm glad the charts were helpful in tracking the process I'm trying to convey. I really do believe this is a much simpler and thorough process to follow to calculate forcing variations (at least the defensive side). Since this is my first post using the chart I think I'm going to leave in the blank chart, but I won't do that for future posts.

Comment 1 -- I need to disagree with you on this one, but perhaps I should explain better. Based on the evaluation of material and activity, you can often determine who is better as I did with the verbal analysis. But there are only two ways to actually play the game -- you are playing to win, or you are playing to draw. If you are better you are trying to keep your lead and win (or not allow equality, since that gets you closer to losing). If you are worse, yes of course you still want to win, but you have to fight for equality first.

Psychologically I think it is much easier to just say "playing to win, or fighting for a draw" to summarize the more lengthy and detailed analysis. Hopefully you are never playing for a loss!

Comment 2 -- This is all related to the primary threat that we determined was dxc1=Q/R#. Yes, promoting the pawn is another threat, and blocking the promotion square with your queen is one way of dealing with it, but promoting the pawn is NOT the most important threat on the board. We're only calculating candidate moves based on the primary threat. It is very important to keep that in mind!!

Comment 8 -- I agree that white is better with the pawn than without it (and thus black should reject this move without further analysis), but practically speaking I don't think it makes it that much easier to win for white. In fact, BECAUSE white's c4 pawn is taking b5 away from the king, black actually has drawing ideas of his OWN to consider.

After (3) 1…d2 2. Qxb2 Ka4 3. Qxd2 black now has the simple 3...Ra3+:

4. Kb2 Ra2+! 5. Kxa2 - stalemate (if white allows RxQ, KxR, the pawn endgame is an easy draw)
4. Kb1 Rb3+ 5. Kc1 Rc3+! wins the pawn (if 5. Qxc3, stalemate; if 5. Kb1/Kb2 Rb3+ repeats). This is the exact same position from the other line. Technically white is winning, but practically it can be very hard to win. Regardless, this position is worse for black than the draw by stalemate.

2. * continuation from the 1st part

8. I address this variation [(3) 1…d2 2. Qxb2 Ka4 3. Qxd2]. "This is almost identical to the variation above, except this time white has an extra pawn, so the evaluation would be slightly better for white – most likely a draw, but winning chances for white." --> I disagree my dear friend. White is MUCH better as Black is not able to win White's pawn and Black is going to lose due to zugzwang motif. It means Black has to pray (quite hard) to save the game!

9. Last two charts (tables): the variation you mention is: 3) 1…d2 2. Qxb2 Ka4 3. Qxd2, but inside the charts (tables) there is just "2...Kb4 3.Qxd2". As far as I caught it - it is a mistake (the same is with the description next to it "Draw likely; white has winning chances" - see the point above).

10. "1…d2 2. Qxd2 Rb1# A clear refutation!". Maybe some small addition is needed. What about: "1…d2 2. Qxd2 (the Queen unpinned the Rook) and now Black plays 2...Rb1# That's a simple and powerful refutation!".

11. "THE CONCLUSION? White's best move is to go for the stalemate." (you may add the BIG text additional phrase)

I think that's all the necessary improvements and suggestions for making this article much better. I hope you will use some of my hints to make it clearer - especially for beginners. Do not make me wrong - even if some corrections are needed, the text is EXTREMALLY good one! This time I had no doubts what you are reffering to and how you are searching all these threats and defences! That means progress my friend (from both of sides!). Thanks!

3. Hello my friend!

I just wanted to inform that I like your corrections - no matter if they take into account my suggestions (recommendations). Of course I am very glad you use these to improve your thoughts and make them more precise. What can I say more?

I am looking forward to your next posts! They are so interesting that I face big problems (difficulties) to wait for next parts! You are doing great job and keep it up my friend! :)

1. Thanks Tomasz. I appreciate all of your thoughtful feedback and I'm really glad that you enjoy this blog series...and I have another one coming today!